TOPLINE:

Severe hospital-treated infections were associated with a higher risk for dementia, regardless of preexisting frailty, comorbidities, or age-related diseases. Participants with severe infections during a 5-year exposure period, including non-frail individuals without comorbidities, had an elevated risk for dementia.

METHODOLOGY:

• Researchers conducted a prospective observational study, including 449,223 participants (mean age, 56.5 years) from the UK Biobank (December 2006-October 2010) and 6106 participants (mean age, 60.6 years) from the Whitehall II prospective cohort study (October 2002-November 2004); all participants were free of severe infections or dementia at baseline.
• Frailty was assessed at baseline using the phenotypic Fried Frailty Scale, which measures unintentional weight loss, low grip strength, exhaustion, slow walking speed, and low physical activity (scored 0-5, with 0 indicating non-frail, 1-2 pre-frail, and ≥ 3 frail), or using a Proteomic Frailty Index on the basis of 25 plasma proteins.
• Comorbidities were assessed using the Charlson Comorbidity Index, and 78 age-related diseases were identified from hospital records.
• Incident severe hospital-treated infections were identified from linked hospital discharge records during a 5-year exposure period from baseline, categorized as bacterial infections, viral infections, pneumonia, gastrointestinal infections, sepsis, and others. This was followed by a 2-year washout period, with dementia ascertainment extending until 2022 for the UK Biobank and 2023 for the Whitehall II study.
• The outcome measured was incident dementia occurring after the exposure and washout periods.

TAKEAWAY:

• The presence vs absence of severe infections was associated with an increased risk for dementia (adjusted hazard ratio [aHR], 1.54; 95% CI, 1.43-1.67) before adjusting for frailty. After adjusting for frailty, no statistically significant attenuation of the infection-dementia association was observed (aHR, 1.49; 95% CI, 1.37-1.61).
• Bacterial infections, viral infections, pneumonia, lower respiratory tract infections, and urinary tract infections were associated with an increased risk for dementia among both non-frail and pre-frail or frail participants compared with those without such infections.
• Severe infections were linked to an increased risk for dementia before (aHR, 1.59; 95% CI, 1.47-1.72) and after (aHR, 1.54; 95% CI, 1.43-1.67) adjusting for comorbidities; similar associations were noted before and after adjustments for age-related diseases (aHR, 1.60; 95% CI, 1.48-1.73, and aHR, 1.52; 95% CI, 1.41-1.65, respectively). Severe infections were associated with dementia in non-frail participants without preexisting comorbidities or age-related diseases, as well as in those who were pre-frail or frail or had at least one comorbidity.
• In the Whitehall II study, participants with severe infections had a higher risk for dementia than those without (aHR, 1.43; 95% CI, 1.02-2.01), which remained similar after adjustment for comorbidities and age-related diseases.

IN PRACTICE:

“This evidence argues against confounding by preexisting frailty or comorbidities and suggests that the association between severe infections and dementia is unlikely to be explained by underlying age-related physical conditions,” the authors of the study wrote.

“Although clear associations between severe infections and dementia persisted after adjustment for frailty in the UK Biobank, results from the Whitehall II study showed weaker, generally null, associations, limiting conclusions,” Janet Janbek, Danish Dementia Research Centre, Department of Neurology, Copenhagen University Hospital-Rigshospitalet, Copenhagen, Denmark, wrote in an accompanying comment.

SOURCE:

The study was led by Pyry N. Sipilä, MD, Department of Public Health, Clinicum, University of Helsinki, Helsinki, Finland. It was published online on April 9, 2026, in The Lancet Healthy Longevity.

LIMITATIONS:

The observational design did not allow for determining cause-and-effect relationships. Dementia identification relied on national hospital and death records, which might have missed milder or undiagnosed cases and delayed the recognition of the disease. The Fried Frailty Scale used to assess frailty may not have been accurate for middle-aged people because the signs of physical frailty were rare at those ages.

DISCLOSURES:

The study received support from the Finnish Medical Foundation, Päivikki and Sakari Sohlberg Foundation, Research Council of Finland, Wellcome Trust, UK Medical Research Council, and National Institutes of Health. One author reported being on the board of directors for the National Academy of Neuropsychology. Another author reported consulting on vaccines for GSK, Merck Sharp & Dohme, and the Finnish Institute for Health and Welfare; participating on a Data Safety Monitoring Board or Advisory Board for two trials; and serving as chairperson for the Finnish national guideline group for cognitive disorders. Janbek reported receiving consulting fees from Eli Lilly and Company.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.